| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesiology and Critical Care Medicine, and Division of Neuroanatomy, Innsbruck Medical University, Austria; and Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts.
Address correspondence and reprint requests to Dr. Philipp Lirk, Department of Anesthesiology and Critical Care Medicine, Innsbruck Medical University, Anichstr. 35, 6020 Innsbruck, Austria. Address e-mail to philipp.lirk{at}i-med.ac.at.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: We investigated lidocaine neurotoxicity in compartmental sensory neuron cultures, which are composed of one central compartment containing neuronal cell bodies and a peripheral compartment containing their axons, allowing for selective incubation. We applied lidocaine ± neuroprotective drugs to neuronal somata or axons, and assessed neuron survival and axonal outgrowth.
RESULTS: Lidocaine applied to the peripheral compartment led to a decreased number of axons (to 59% ± 9%), without affecting survival of cell bodies. During axonal incubation with lidocaine, the p38 mitogen-activated protein kinase inhibitor SB203580 (10 µM) attenuated axonal injury when applied to the axon (insignificant reduction of maximal axonal distance to 93% ± 9%), but not when applied to the cell body (deterioration of maximal axonal length to 48% ± 6%). Axonal co-incubation of lidocaine with the caspase inhibitor z-vad-fmk (20 µM) was not protective.
CONCLUSIONS: Whereas inhibition of either p38 mitogen-activated protein kinase or caspase activity promote neuronal survival after LA treatment of dissociated neuronal cultures, axonal degeneration induced by lidocain (40 mM) is prevented by p38 MAP kinase but not by caspase inhibition. We conclude that processes leading to LA-induced neurotoxicity in dissociated neuronal culture may be different from those observed after purely axonal application.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
However, all currently used LAs are, to some extent, neurotoxic (3–5). Neurotoxicity elicited by the commonly used LA lidocaine remains a considerable obstacle in patient safety during regional anesthesia (5,6). Direct LA toxicity may already be discernible at clinically relevant doses (7,8), leading to transient, or even permanent, neurological damage (9). The incidence of permanent or transient neurological damage subsequent to peripheral nerve block is commonly considered less than after neuraxial blockade. For example, Lee et al. (10) demonstrated in a closed-claims analysis that peripheral nerve blocks associated with nerve damage accounted for only 5% of closed regional anesthesia claims. However, it is notable that, in a prospective survey involving regional anesthesia procedures in more than 158,000 patients, the incidence of neuropathy after peripheral nerve block was as frequent as after central blockade, with a considerable number of patients suffering from permanent nerve damage (11). Indeed, it has been suggested that local tissue damage during peripheral nerve block is a considerable concern (6).
The pathways leading to neuronal damage after regional anesthesia are incompletely understood. Therefore, knowledge of pathogenic mechanisms will be essential to devise strategies aimed at further decreasing the incidence, or at least minimizing the extent, of LA-induced neurotoxicity. Although toxicity of both conventional (3,12,13) and investigational (14) LAs has been examined in vitro, these experiments were conducted in dissociated cultures of primary sensory neurons (12,14) or immortalized cell lines (15). These investigations suggest that apoptotic pathways, such as loss of mitochondrial membrane potential and caspase activation, may be involved in LA-induced neurotoxicity (5,12,13). However, this approach may not accurately reflect the in vivo situation for peripheral nerve block, because in dissociated cell cultures the entire cell (soma and processes) is immersed in the experimental substance. In contrast, during clinical practice, peripheral conduction block is elicited by applying the LA to the axon, sometimes far from the cell bodies, which are assembled in the dorsal root ganglion (DRG). For example, during axillary plexus blockade, LA is applied to the peripheral nerve some 20 cm distal from the DRG. Therefore, even though dissociated neuron cultures are valid models in which to test the general effects of neurotoxic or protective drugs, they do not offer the spatial resolution necessary to closely simulate in vitro the application of LA to nerves to achieve a conduction block.
Recently, the involvement of stress-activated protein kinases such as the p38 mitogen-activated protein kinase (MAPK) was demonstrated after incubation of neuronal cultures with the LAs lidocaine and tetracaine (4,12,16). One potentially protective intervention is thus the co-injection of an inhibitor of the p38 MAPK with an LA (12). In dissociated DRG neuron cultures, specific activation of the p38 MAPK, but not other related stress-activated protein kinases such as the c-jun N-terminal kinase, has been described in response to LA treatment (17). However, it is not known whether p38 MAPK inhibitors act directly at the neuronal axon or indirectly by interfering with pathways in the cell body.
The compartmented model of neuron cultures has been proposed as a means to test the influence of neuroprotective or neurodegenerative interventions selectively for soma and axon (18,19). The overall goal of this series of experiments was to describe the compartmented neuron culture model as a possible tool to investigate, in greater spatial detail, mechanisms leading to axonal neurotoxicity after regional anesthesia. Specifically, it was the aim of the present study to test in compartmental neuron cultures of primary sensory neurons the hypothesis that lidocaine-induced axonal toxicity is mediated by processes confined to the axon. As axons may undergo selective degeneration without evidence of apoptosis upon injury (20), we also sought to preliminarily determine whether apoptosis-related pathways such as caspase activation are involved in lidocaine-induced axonopathy.
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Drugs
Unless stated otherwise, drugs were purchased from Sigma Aldrich (Vienna, Austria, or St. Louis, MO). The pH of the lidocaine stock solution 1 M was 4.65 (in dimethylsulfoxide). The pH of the final solution added to medium was 7.38 for lidocaine 40 mM. Control cultures (or compartments) were incubated with RPMI (Roswell Park Memorial Institute) medium equal in dimethylsulfoxide content and osmolality to experimental cultures.
Primary Sensory Neuron Culture
Neurons were acutely harvested from adult (8–9 wk old) female Sprague-Dawley rats, which were euthanized by CO2 narcosis according to the institutional protocol (Animal Committee of the Austrian Federal Ministry of Education, Science and Culture, Vienna, Austria). DRGs were de-sheathed and incubated in collagenase 5000 U/mL for 60 min at 37°C, followed by 15 min incubation with 0.25% trypsin/EDTA. After dissociation in RPMI medium containing 10% horse/5% fetal bovine serum, neurons were plated in RPMI medium supplemented with N2 additives (1:100) and antibiotics (penicillin, 1000 U/mL; streptocmycin, 1000 µg/mL; and amphotericin B, 25 µg/mL in 0.85% saline, all purchased from Invitrogen, Vienna, Austria). Thirty-five millimeters Falcon tissue culture dishes were coated with poly-d-lysine at a concentration of 0.1 mg/mL in distilled H2O and laminin at 7 µg/mL in RPMI solution, and cell cultures were kept at 37°C in a humidified atmosphere containing 5% CO2.
Compartmental Neuron Cultures
Compartmental cultures were prepared using the model described by Campenot (18), seeding neurons into a central chamber surrounded by a Teflon divider (Tyler Research Instruments, Edmonton, Canada) sealed to the bottom of the tissue culture dish (Falcon) with silicon grease (Dow Corning). For a scheme of the experimental setup refer to Figure 1. Briefly, 35-mm culture dishes were coated with poly-d-lysine and laminin as described above, and a series of parallel tracks in the substrate was made by scratching the culture dish surface with a pin-rake (Tyler Research Instruments). Subsequently, a three-chambered Teflon divider was seated on top of the tracks, crossing the scratches at right angles, and sealed to the dish with silicone grease. Dissociated neurons were plated into the central compartment and maintained in RPMI medium supplemented with N2 additives and antibiotics. Culture conditions were modified from those described by Finn et al. (20). In brief, DRG neurons plated into the central chamber were allowed to adhere for 5–7 days, and each neuron grew one major process subsequently designated as "axon." Nerve growth factor (100 ng/mL) was added to medium (RPMI supplemented with N2 additives and antibiotics) in distal compartments to promote elongation of neurites. Medium was changed every 2–4 days, and cultures were maintained in a humidified atmosphere at 37°C. After 7 days, axons had extended through the silicon grease into the two distal compartments, and images were taken at 20x magnification with a Zeiss Axiovert 100M microscope (Zeiss, Germany) with a digital camera (Spot RT) connected to a PC. Subsequently, the compartmented cultures were subjected to different experimental protocols, and analysis of maximal axonal distance was performed with Metamorph software (version 6.2r5, Visitron Systems, Munich, Germany).
|
We determined maximal axonal distance, defined as the distance from the barrier which separates the central from the peripheral compartments to the most distant part of the axon. After images were taken of neurites by phase-contrast microscopy (20x magnification, Day 7), 40 mM lidocaine was added to the compartments. In five cultures, images of the investigated neurites were taken every 10 min for 1 h to document the initial morphologic changes induced by lidocaine at the axonal level. In preliminary experiments, we had ensured that both the p38 MAPK inhibitor, SB203580, and the caspase inhibitor, z-fad-fmk, added to cell cultures did not interfere with axonal outgrowth (data not shown).
Pharmacologic Interventions
Lidocaine Neurotoxicity in Dissociated Primary Sensory Neuron Cultures
Our group previously described that incubation of neuronal cultures with lidocaine results in a decrease in cell number, and that concomitant inhibition of the p38 MAPK results in improved survival (17). As a next step, we sought to further determine whether caspase activation, as a hallmark of apoptosis, contributes significantly to lidocaine-induced cytotoxicity in dissociated neuronal cultures. This was achieved by incubating neuronal cultures with 40 mM lidocaine for 24 h, with or without the addition of a nonspecific caspase inhibitor known to predominantly block caspases-1 and -3, z-vad-fmk (20 µM). We determined neuron number in cultures by counting the number of neurons along two passes through the diameters of each well at 20x magnification. DRG neurons typically have a large and distinctly spherical cell body with a clearly visible nucleus, and are thereby easily distinguished from nonneuronal cells, such as glial cells.
Lidocaine Neurotoxicity in Central and Peripheral Compartments
Experimental treatments of the compartmented cultures included exposure of either the cell bodies or the neurites to 40 mM lidocaine for 24 h. Lidocaine was added to the peripheral or central compartments after neurite extensions had been recorded and number of axons or cell bodies, respectively, had been determined 7 days after plating. After 24 h of lidocaine treatment, images were taken of axons in the peripheral compartments for growth measurements, and axons and cell bodies were counted in both peripheral and central compartments.
Neuroprotection During Axonal Lidocaine Incubation
Next, we sought to determine whether the neuroprotective drugs act locally at the axon or via the neuron's soma. To this end, we incubated the neuronal axon in the peripheral compartment with lidocaine, either with or without the p38 MAPK inhibitor SB203580 (10 µM). In a second set of experiments, the neuronal axon in the peripheral compartment was again incubated with lidocaine, but this time, the p38 MAPK inhibitor was simultaneously applied to the cell body. Neuronal survival and maximal axonal length were used as outcome measures. We did not apply lidocaine combined with SB203580 to the central compartment because this interaction can best be studied in an acutely dissociated DRG cell culture model and has been described before (17). In a third set of experiments, we applied lidocaine to the peripheral compartment, and added the nonselective caspase inhibitor, z-vad-fmk, to either the central or the peripheral compartment.
Statistics
Data from all neurons in each culture were pooled with equivalent cultures at the same time point and analyzed by one-way ANOVA with Prism statistical software (GraphPad Prism 3.0). Experiments were performed in four to six independent cultures, and average axon number analyzed in each culture was 20. Data were normalized to percent values (controls = 100%), and results are given as mean ± standard deviations (sd). Statistical significance was assumed at P < 0.05. We used a scaled mean length per culture dish as a unit of analysis, and therefore no weighting by culture was necessary. As a representative example, lidocaine results on culture level data were comparable to those performed on the axon level, and were consistent with other analysis methods, including fixed effect regression models with culture modeled as a repeating variable (P = 0.0172).
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Lidocaine Incubation in Compartmented Neuronal Cultures
First, we sought to determine morphological sequelae. After incubation with 40 mM lidocaine into the peripheral (axonal) compartment, the first signs of axonal degeneration, such as retraction of distal neurites, were observed within 10 min. In addition, lidocaine caused consistent fragmentation of neurites similar to the distinctive "dying-back" morphology described by Laser et al. (21) and Klimaschewski et al. (22), beginning distally and proceeding proximally (Fig. 3).
|
The addition of 40 mM lidocaine to the peripheral compartment (n = 6), as well as addition of 40 mM lidocaine to the central compartment (n = 6), significantly reduced the axon number from baseline values (100%) to 59% ± 9% or 58% ± 12% after 24 h, respectively (Fig. 4a). Moreover, addition of 40 mM lidocaine into the central compartment (n = 6) significantly reduced the number of neuronal cell bodies from control values (at Day 7) to 60% ± 8% after 24 h, whereas addition of 40 mM lidocaine into the peripheral compartments (n = 6) had no such effect (Fig. 4b).
|
Lidocaine-Induced Neurotoxicity is a Localized Process
The addition of 40 mM lidocaine into the central compartment (n = 4) severely reduced maximal axonal distances to 43% ± 10% compared with baseline after 24 h. Incubation with 40 mM lidocaine in the peripheral compartment (n = 4) also significantly decreased maximal axonal distances as measured on Day 7 after seeding to 53% ± 6% 24 h later. Co-incubation of these cultures with the p38 MAPK inhibitor SB203580 at 10 µM in the peripheral compartment (n = 4) showed an attenuation of neurotoxic effects, in that the maximal axonal length was 93% ± 9% after 24 h. Addition of 10 µM SB203580 into the central chamber (n = 4) had no protective effect (maximal axonal distance 48% ± 6%), indicating that the toxic effect induced by lidocaine is localized to the axon itself (Fig. 5).
|
Caspase Inhibition does not Reduce Axonal Toxicity in Compartmented Neuron Cultures
In compartmented neuronal cultures, controls of this experimental setup showed an increase of maximal axonal distance from 100% at Day 7 to 133% ± 5% at Day 8. Co-incubation of compartmented cultures with 40 mM lidocaine in the peripheral compartment and z-vad-fmk at 10 µM in the peripheral compartment (n = 3) showed that peripheral inhibition of caspase activation had no significant effect on the axonotoxicity induced by lidocaine. The cultures treated with 40 mM lidocaine alone in the peripheral chamber had a maximal axonal distance 60% ± 9%, whereas addition of z-vad-fmk resulted in a maximal axonal distance of 77% ± 8% at Day 8. Moreover, simultaneous incubation of cultures with 40 mM lidocaine in the peripheral and 10 µM z-vad-fmk in the central compartment showed no significant protective effect on axon outgrowth (maximal axonal distance 72% ± 9%, Fig. 6). Results are summarized in Figure 6.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We chose to use the model of compartmental culture described by Campenot (18,19), which allowed us to apply the drugs being tested selectively to either axon or soma. Finn et al. (20) have shown in an identical compartmental model that a neuron's response to injury varies according to the site of injury. Specifically, axons are able to undergo a process independent of the apoptotic neuronal cell death, referred to as selective axonal degeneration.
The mode of peripheral axonopathy observed by us after incubation with lidocaine is consistent with a very rapid activation of destructive pathways in dissociated cell cultures (23). We note that the morphologic appearance of axons undergoing degeneration is reminiscent of recent findings by Laser et al. (21) and Klimaschewski et al. (22), who demonstrated immediate growth arrest of axons grown from neonatal explant cultures or dissociated adult neurons in response to treatment with proteasome inhibitors. This acute effect was followed by degeneration of the distal axons and fragmentation of neurites resembling axon degeneration in human neuropathies (24), and in animal models of neurological diseases (25), such as gracile axonal dystrophy (26).
LAs have been shown to elicit neuronal degeneration, including growth cone collapse in primary sensory neurons (27). However, some studies reporting these effects were performed in dissociated cultures from neurons harvested from chick embryos, which makes direct comparison of data difficult (27,28). Kanai et al. (29) could show that application of lidocaine to dissociated cultures of mouse primary sensory neurons resulted in reversible inhibition of anterograde and retrograde axonal transport within <10 min. Similar to the present study, an inhibitory effect of lidocaine upon the growth rate of DRG neurites was demonstrated after incubation with small doses (30 µM) of lidocaine for 30 min (29). Correspondingly, larger doses of lidocaine (30 mM), comparable to those used in our study, have been reported to lead to neurite destruction (30). Therefore, the general adverse effects of lidocaine upon axonal structures, and in particular the rapid mode of axonal degeneration demonstrated in the present study, correlate well with observations in previous literature (27–30).
We demonstrated that LA applied to the central (soma-containing) chamber of compartmental cultures led to neuronal degeneration, with loss of both cell bodies and axons. This result corresponds well with previous reports of neurotoxicity in dissociated cultures (12). In contrast, if only the axon is incubated with lidocaine, cell body survival is not compromised, even though the morphology of peripheral axons suggests a highly toxic effect. These observations strongly indicate that in vitro experiments connected with neurotoxicity and neuroprotective factors should optimally be conducted in a model that allows the environments of the soma and the axon to be controlled separately. This is imperative because, in contrast to other cell types, neurons appear to feature at least two distinctive self-destructive programs: apoptosis involving the entire cell and selective axonal degeneration (24). The latter is confined to the neuron's periphery and is molecularly distinct from the apoptotic cascade, e.g., lack of caspase activation (24). For this reason, caspase inhibitors, effective in attenuating apoptosis, may be ineffective in alleviating selective axonal degeneration (20). Raff et al. (24) postulate that the physiological role of selective axonal degeneration is to eliminate undesired axonal branches during development and to conserve resources after neuronal injury.
In our experiments, we applied lidocaine to the axonal compartment and added the p38 MAPK inhibitor, SB203580, either to the soma or the axon. Inhibition of the p38 MAPK has been shown to confer protection from lidocaine-induced neurotoxicity in vitro and in vivo (12). We demonstrate that application of p38 MAPK inhibitor to the site of injury, i.e., axon, resulted in substantial attenuation of LA-induced neurotoxicity. To exclude the possibility that the p38 MAPK inhibitor would be transported to the cell's soma, where it would indirectly exert its neuroprotective effects, we applied SB203580 to the neuronal soma while incubating peripheral axons with lidocaine and found no protective effect. This is notable, because p38 MAPK is typically thought to act via modulation of nuclear factors such as transcription factors, or by activating lipoxygenase pathways. Here we provide evidence that p38 MAPK may be involved in degenerative processes confined to the neuron's periphery.
Further, we sought to test whether pathways of apoptosis such as activation of caspases are relevant effectors of lidocaine-induced axonopathy. Even though our experiments provide only indirect evidence that will need to be confirmed in an in vivo study, we note that inhibition of caspase activation is neuroprotective in dissociated, but not in compartmented, neuron cultures. Specifically, we note that dissociated cultures were protected by the caspase inhibitor z-vad-fmk against lidocaine-induced neurotoxicity, which corresponds well with previous literature (14). However, inhibition of the same pathway was not protective at the axonal level during selective incubation of the axon with lidocaine. Together, our findings suggest that the neurotoxic pathways are different, depending upon the site of injury to the nerve. We hypothesize that when the entire cell is immersed in LA, apoptosis is elicited (17). In contrast, when the axon is the site of injury, selective axonal degeneration may be the process responsible for lidocaine-induced peripheral axonopathy. These findings are of considerable clinical relevance in the quest to identify mechanisms of LA-induced neurotoxicity. We propose that, for in vitro investigations of axonal neurotoxicity, one should resort to compartmented neuronal cultures, whereas situations in which the entire cell is exposed to LA, such as in neuraxial anesthesia, can be simulated with a regular dissociated cell culture model.
Finally, some limitations of our study design should be briefly mentioned. We attempted to create uniform culture conditions for all experimental groups, but acknowledge that some variation existed between cultures concerning their growth rate at the time of incubation (as evident when comparing Figs. 5 and 6). We harvested mature neurons from rodent DRGs, thereby severing their axons, and subsequently applied drugs to the neurons while they were regenerating. This may be relevant as it could amplify the adverse effects of LAs (31).
In summary, we describe compartmental primary sensory neuron cultures as potentially useful to determine in greater detail pathways involved in LA-induced axonotoxicity. We note neuroprotective effects of a specific p38 MAPK inhibitor, but not caspase inhibitors, when co-administered with lidocaine to the axon of a primary sensory neuron. We conclude that LA-induced neurotoxicity may be mediated via different mechanisms, depending on whether the entire cell, or only the axon, is incubated.
|
Footnotes |
|---|
Supported by the National Institutes of Health, Bethesda, Maryland (Research Grant No. GM64051 to P. Gerner).
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
